OCRRouter

Marker vs dots.ocr

Pick Marker if you want PDF to Markdown with a CPU option, a hosted API, or RAG chunks, and your company is under $5M or will buy a licence. Pick dots.ocr if you want an MIT-licensed model on a CUDA GPU that returns layout boxes, tables as HTML and formulas as LaTeX in one pass. The main split is licence and deployment: Marker has a restricted weight licence but runs on CPU and has a hosted API, while dots.ocr is self-host only and needs a GPU.

Markerdatalab/markerdots.ocrrednote/dots-ocr
TypeOCR APIOpen source
Price14 credits / page2 credits / page
Per 1,000 pages$14.00$2.00
Typical speed2.5 s / page1.5 s / page
Best forBooks and research papersMultilingual document parsing
InputsPDF, images, DOCXPDF, images
Languages90+ languages100+ languages
Max pages per request1,000500
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxes✓ Yes✓ Yes
ConfidenceNoNo
FormsNoNo
Multilingual✓ Yes✓ Yes
On OCRRouterAvailableAvailable

Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.

Which to choose

Choose Marker if

  • You need to run on CPU or Apple Silicon: fast mode uses about 18 MB of ONNX weights and runs at 23.6 pages per second, though it scores 66.6 on olmOCR-bench.
  • You want a hosted option: Datalab's API includes about 5,000 free pages per month, with paid plans on its pricing page.
  • You need RAG chunks or JSON block trees, and you work with books and papers that contain equations.

Choose dots.ocr if

  • You want MIT-licensed code with no revenue cap on commercial use of the code. The repository also includes a separate dots.ocr license agreement file.
  • You have a CUDA GPU and want one 3B model that returns layout boxes, categories and text, with tables as HTML and formulas as LaTeX.
  • You work with multilingual PDFs or mixed table and text documents, or you need charts and diagrams converted to SVG with dots.mocr-svg.

Key differences

Markerdots.ocr
LicenceCode Apache-2.0. Weights under a modified AI Pubs Open Rail-M license: free for research, personal use and startups under $5M.MIT, plus a separate dots.ocr license agreement file in the repository.
How you run itLocal on GPU, CPU or Apple Silicon (MPS), or Datalab's hosted API. vllm mode needs an NVIDIA GPU.Self-host only, via vLLM or Hugging Face Transformers. Single CUDA GPU with vLLM is the documented path. No VRAM figure stated.
PriceFree self-hosted within the weight licence. Hosted API has about 5,000 free pages per month.Free, self-hosted. You pay for compute.
Output formatsMarkdown, HTML, JSON, chunks, LaTeX, images.JSON, Markdown, HTML, LaTeX, bounding boxes, SVG.
olmOCR-bench (authors' own numbers)83.7 in vllm mode. 66.6 in CPU fast mode.79.1 for dots.ocr. 83.9 for the newer dots.mocr (v1.5).
Model sizeNot published3B parameters (1.7B LLM backbone).
Known failureWeight licence blocks free commercial use for companies above $5M. CPU fast mode scores lower.Runs of ellipses or underscores can make output repeat endlessly. Complex tables and formulas remain difficult. Occasional parsing failures.

Sources

What each one is

Marker

Hosted version of the open-source Marker and Surya models. Strong on books and papers.

Marker details

dots.ocr

Single vision model from rednote that does layout detection and content recognition together.

dots.ocr details

Try both on your own file

Run the file once with each model and compare the output.

PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?

Use one, fall back to the other

{"model": "datalab/marker","fallbacks": ["rednote/dots-ocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Marker alternativesdots.ocr alternativesBest OCR models in 2026